What is being presented?
The Royal Marsden research team, led by Rosalind Eeles,
presented their findings from the Barcode 1 study at the ASCO 2024 meeting.
This prostate cancer test examines men's DNA for variations associated with a
diagnosis of prostate cancer. Unlike traditional methods that rely on mutated
DNA from tissue samples of the prostate, this test examines DNA in normal
tissue—specifically, the DNA that patients are born with. The test is
non-invasive, relatively inexpensive, and can be performed at home, offering a
high degree of convenience. It is proposed as an alternative to the more
conventional PSA blood test.
The majority of the DNA changes detected in these tests are
SNP or single nucleotide polymorphism. Associations are established by
screening for millions of variations in the DNA code, to see whether variations
in DNA in specific locations are associated with prostate cancer. As there are millions
of DNA sites being tested then DNA from 100,000s of men with and without prostate
cancer need to be examined to try and establish these associations. Because
there are millions of hypotheses being tested at the same time, then typically
a threshold probability of 10-8 may be employed to establish an association,
which is roughly equivalent to the typical probability threshold of 0.05 when a
single hypothesis is tested. Using this higher threshold aims to filter out
spurious associations. Although it is not understood what effect these genetic
variations have on a cell, robust associations with disease can be established.
These proposed associations then need to be tested on a separate cohort to
validate the findings and the proposed risk model. In this study the risk model
or polygenic risk score for prostate cancer was based on variations in 130 specific
nucleotides which were found previously to be associated with prostate cancer.
The test is a novel concept in cancer screening as it aims
to find patients who are genetically at a higher risk of developing prostate cancer,
and therefore cancer could be detected at an earlier stage. The test may facilitate
some patients being diagnosed with prostate cancer who would not be detected
through PSA and MRI, as patients with higher risk disease were diagnosed despite
having normal PSA values or MRI prostate scans.
So how does this compare to PSA testing?
This type of test would be a one-off test as the DNA which
we are born with does not need to be retested. This could be viewed as both an advantage
and a disadvantage. Not needing to repeat a test makes it less expensive and
time consuming for screening purposes. The highest diagnostic yield is usually
accrued when a test is first undertaken, as the yield progressively diminishes with
repeat testing. However, repeating a test in patients allows screening to be a
continuous process and can allow disease progress to be monitored. It is known
from the European Randomized study of Screening for Prostate Cancer that
prostate cancer screening is not a single shot strategy as patients in this
study were diagnosed with prostate cancer steadily over 16 years using PSA as
the primary screening tool.
Is it better than PSA testing?
I don’t think we can tell from this study though further
clarification may arise from the full study protocol. The picture is incomplete
as only patients in the top 10% of the polygenic risk score had further evaluation.
What we don’t know is from the 90% of patients below this threshold, whether they
had elevated PSA and whether further evaluation would have led to the diagnosis
of additional cancers. We don’t know the false negative rate for polygenic risk
score. I.e., patients who did have prostate cancer but unremarkable polygenic
risk score. We also don’t know how the test behaves at differing cut off
thresholds i.e., top 20% or 30% to see the differing trade-offs between specificity
and sensitivity, to consider the optimal threshold for screening purposes.
We can make some inferences here by comparing the results
with another UK screening study- the Protect study which is based on PSA
testing. Here a threshold of PSA >3.0 was used to determine study entry and
further patient evaluation. Coincidentally, this also represents the upper 10%
strata for PSA test results in the study population.
In the Barcode 1 study 2.8% of patients who consented and
40% who were biopsied were diagnosed with prostate cancer. By comparison, in
the Protect Study, 2.9 % who were recruited and 39% who were biopsied were diagnosed
with prostate cancer. However, the Protect study was carried out in the pre-MRI era, where the accuracy of biopsy was less. Typically 60% of patients who now undergo
a biopsy for prostate cancer, detected by elevated PSA and localised by MRI, are
diagnosed with prostate cancer.
So, in conclusion it probably won’t replace PSA testing, but
it does offer an additional tool that may complement existing strategies. PSA
testing will probably generate a higher diagnostic yield. This isn’t surprising,
as according to our current understanding, only a minority of patients have a hereditary
cause for their prostate cancer. The patients detected by this method may represent
a different cohort of patients. In the fullness of time, we may find that their
disease takes a different course, and the risks benefit of up-front treatment
may be more favourable. One of the reasons that screening for prostate cancer
is not yet currently employed is that the benefits of early treatment are
marginal and not clearly established. It may be true that we already have screening of sorts with the ease of availability and frequency with which PSA
tests are already conducted.
So, what’s the future of Prostate Cancer Screening?
Screening could take several forms. A strong candidate is up-front MRI of the prostate, which can cost as little as £145 if they are bi-metric scans which omits contrast enhancement. But blood tests, saliva tests, and urine tests could be employed. Another possibility which may be relatively under evaluated are semen tests, as this tests a direct product of the prostate gland. A previous study evaluating semen found that the concentration of citrate in the semen could be used to predict prostate cancer to a greater degree than through the PSA blood test.
Seminal citrate is superior to PSA for detecting clinically significant prostate cancer